It errors on the first line read, i.e. the file is either empty or does not exist. I would not write it in ascii, at least the code requests a binary file and reading a non-binary file can really mess with the read quantity (if it will work at all).
It should write out the DIM file if it converged, so I would check the out file whether it really did converge. Please try again. Kind regards Nick 2013/9/26 Ian Shuttleworth <[email protected]> > The actual error I'm getting is: > > > > At line 89 of file local_reinit.f (Unit 11 "siesta.DIM") > Traceback: not available, compile with -ftrace=frame or -ftrace=full > Fortran runtime error: End of file > > > Either it cant see the file, or it doesnt understand what's in the file. > The DIM file is named siesta.DIM (default). > I used DIM in ascii previously because I had trouble actually getting > the file to write - if you do a CG steps = 0 run, I found the DIM file > didnt always write so I wrote it myself in plain text. I assumed the > non-writing if CG=0 was a quirke of my implementation. I was going to > do the same again now. > > > On Thu, Sep 26, 2013 at 11:42 AM, Nick Papior Andersen > <[email protected]> wrote: > > I cannot seem to find the problem. In siesta-3.2-pl3 the code which > writes > > the DIM file is (as you say) binary, and is performed in plcharge.F > > In Denchar the file which reads it is local_reinit.f and that also > assumes > > it to be in binary format. So no mismatch here. > > > > I just checked in siesta-3.1 and that one has the same way (everything in > > binary). > > > > Are you by any chance using a very old version of denchar? > > And if not, are you sure the problem is the DIM file? > > > > Kind regards Nick > > > > > > 2013/9/26 Ian Shuttleworth <[email protected]> > >> > >> Siesterers > >> > >> I am trying to use DENCHAR within SIESTA-3.2-pl3 > >> > >> Denchar is rejecting the DIM file generated by the SIESTA code > >> generated in the same distribution > >> > >> Looking at the DIM file, format is binary, and inspection of DENCHAR > >> seems to require non-binary format. This is a new problem, I used > >> DENCHAR successfully in earlier versions. DIM files previously were > >> plain text. > >> > >> Has anyone else encountered this problem? Any ideas for a fix? > > > > > > > > -- > To understand the middle east is to understand blood, oil, and milk. > > In that order. >
